Joffre’s plan had been to concentrate superior forces in the French 4th Army sector and attack the weak German centre, break through and cut off the German right wing. The 3rd Army would protect the 4th Army’s right flank. Upon learning that the French were advancing in its sector, the German 5th Army immediately attacked. Neither the German 5th Army nor its right-hand neighbour, 4th Army, were concentrated for combat; one corps in 5th Army and two in 4th Army were a day’s march behind the leading corps. Therefore, Joffre’s plan succeeded: five corps of the French 4th Army faced three corps of German 4th Army.
In the French 4th Army area of operations, II CA was stopped by a German brigade at Bellefontaine and a German regiment at Robelmont. This left the right flank of the 3 DIC exposed and the division was destroyed, as was the 5th Colonial Brigade to its left. XII CA, which was between the Colonial Corps and XVII CA, had been practically inert. In failing to reach its objectives, it had uncovered the flanks of both neighbouring corps. 33 DI of XVII CA had also been enveloped and destroyed. 34 DI, the second division of this corps, disintegrated. On the 4th Army left flank the XI CA had failed to inflict a decisive defeat on an outnumbered and exposed German division. Indeed, the second brigade of that German division had crushed three French brigades and forced XI CA to retreat.
On the left flank of the French 3rd Army, all four infantry regiments of 8 DI had taken such heavy losses that they were no longer combat-effective and one brigade of 7 DI had been wiped out. The next division on the right, 9 DI, had been defeated by 1100 hours. 10 DI had been ordered to retreat. Of the three VI CA divisions, 12 DI had been forced to retreat, 42 DI had retreated even further, and 40 DI had been smashed. All of the remaining divisions in both French armies had broken contact with the Germans and retreated during the night to escape destruction
Due to superior German reconnaissance and troop-leading procedures, the Germans had almost always seized the key terrain and deployed first. The German infantry and artillery gained fire superiority. French units were often caught in march column and the French artillery was slow in supporting the infantry. German units were able to manoeuvre on the battlefield; the French were not. The German reserve infantry units had proved themselves to be at least as effective as the French active-army units. The German units had proven themselves to be superior to the French in every regard.
The corps and army HQ on both sides lost control of their units. Many of the German units could have pursued on 22 August but never received orders to do so. The situation was unclear to German leaders at division level and above, and they preferred ‘safety first’. When the extent of French defeat became evident in late morning on 23 August, the French were out of range and recovering from their defeat.
French casualties overall had been three or four times higher than those they had inflicted on the Germans. Nevertheless, German losses had been significant and these, as well as the exertions of hard marching, combat and a night spent digging defensive positions, had worn down the German soldiers, who on 23 August were physically and mentally exhausted.
Army HQs, 22 August 1914
German 5th Army
Crown Prince Wilhelm wrote that the leadership of AOK 5 knew intuitively that they had won the battle on the entire front. But the tremendous tensions that arose during the day from the continual stream of messages, issuing orders, and the associated feelings of anxiety and satisfaction, had given rise to grave concern. The massive casualties suffered tempered the elation of victory. The intensity of the battle and lack of information concerning the French – AOK 5 did not know how badly the French had been hurt – left the possibility that the French might attack on 23 August. AOK 5 did not know the situation in the 4th and 6th Army sectors. It also wanted to await the arrival of five Landwehr brigades from the Metz garrison. AOK 5 was concerned that the French could have already massed a large force at Verdun to attack the 5th Army left flank. AOK 5 reported to OHL that it had won the battle, but that did not intend to advance on 23 August. The German 5th Army clearly felt that it had not decisively defeated the French. Given the high casualty rate in 5th Army had suffered, the possibility of a French retour offensif had to be guarded against.
French 3rd Army
The French 3rd Army, which opposed the German 5th Army, reported around noon on 22 August that it had taken heavy casualties but its situation was good and that the enemy had suffered just as severely. The GQG liaison officer to 3rd Army reported at 2200 that the 3rd Army commander, Ruffy, had briefed him on the situation. Ruffey said that the 3rd Army held the same line as it had that morning, from Virton to Joppecourt. The army had made only limited progress because 7 DI of IV CA and 9 DI of V CA had been surprised and taken serious losses. The situation had been stabilised by the French artillery, which had established a ‘significant superiority’ over the German artillery. The LNO praised the ‘calm and coup d’oeil of General Ruffey’. The 3rd Army attack had run into enemy troops ‘in solid defensive positions’. The enemy seemed to have intended to mask Longwy while turning the army right flank, that is to say, VI CA and 7 DC. Ruffey complained that the 7 DC had not covered the VI CA right flank. In addition, Ruffey complained that 54 DR had not covered the right flank of VI CA, as he had ordered. Closer examination reveals that the division’s leading elements were at Spincourt, exactly where Ruffey had ordered them to be, about 15km behind 40 DI. Ruffey’s mission for the reserve divisions had been to cover a German attack from Metz to the south. Only after the battle did Ruffey contend that he expected them to operate towards the east.
On 22 August 3rd Army reported that it had been engaged with the German VI AK, XVI AK and a brigade or a division of XIII AK. That the army (seven divisions) had taken such punishment at the hands of, at most, five German divisions would have been alarming.
After the war Engerand said the 3rd Army had fought outnumbered seven divisions against ten: Engerand was explicitly saying that the German reserve divisions were the equals of French active divisions. He maintained that the French 3rd Army had inflicted such heavy casualties on the German 5th Army that it was immobilised for 24 hours. This was incorrect. For most of the day the odds were practically equal, fourteen French active brigades against thirteen German active brigades and three reserve brigades. In the afternoon four more German reserve brigades appeared. At the end of the day many, perhaps most, of the German units were still quite capable of offensive operations; none of the French brigades were. What had been immobilised was not the German manoeuvre units but the mindset of German 5th Army HQ.
German 4th Army
For the entire day on 22 August the German 4th Army was in a precarious position. The army right flank was in the air. There was a gap between XVIII AK on the right and XVIII RK in the centre, leaving XVIII AK isolated. XVIII RK and the corps on the left, VI AK, faced superior numbers of enemy forces. It was absolutely necessary to protect the right flank and concentrate the army. At 1400 AOK 4 ordered VIII RK to push towards XVIII AK as far as possible. VIII AK was to swing south towards XVIII AK. 4th Army sent situation reports to OHL, AOK 3 and AOK 5. At about the same time, an order arrived from OHL, instructing AOK 4 to move its right flank units to the west, south of Dinant, to cut off the French forces facing the German right wing. This order bore no relation to the situation on the ground; AOK 4 disregarded it and VIII AK maintained its march south.
French 4th Army
Langle de Cary reported to GQG on the evening of 22 August that the results of the day’s combat were ‘not very satisfactory’. There had been serious reverses at Rossignol and Ochamps, which negated the successes gained by XI CA and XII CA. Nevertheless, at 2330 he ordered the Colonial Corps, XII CA and II CA to hold their present positions while XVII CA and XI CA resumed the attack to the north, supported by the newly arrived IX CA and 60 RD.
At 0130 on 23 August 4th Army sent a sobering report to GQG. In II CA the 3 DI was in good shape at Meix devant Virton, but the 4 DI had been thrown out of Bellefontaine and had been ‘sorely tried’. The 3 DIC and 5th Colonial Brigade had also been ‘sorely tried’. XII CA was in good shape and had not even engaged its corps artillery, but was falling back. XVII CA was in poor condition, 33 DI had lost its artillery, 34 DI had been thrown back. XI CA had pulled back to the Semois.
More French Fantasies
Both Ruffey and Engerand asserted that, had only the 54 DR and 67 DR attacked, and the 7 DC acted energetically, then the French 3rd Army would have turned the flank of the German 5th Army, perhaps cut it off from Metz, and the French would have won a great victory, which, Engerand maintained, would have had ‘the greatest consequences for the entire front’, and doubtless the entire German position in Belgium would have been shaken. At the very least, the right flank of the 3rd Army would have been secured.
That all this did not come to pass, according to Engerand, was due solely to GQG’s mishandling of the Army of Lorraine. GQG changed the organisation and commanders of the Army of Lorraine and removed it from 3rd Army control – without notifying 3rd Army – on the very day it launched the 3rd and 4th Army offensives. Simultaneously, GQG gave the Army of Lorraine a purely defensive mission, instructing it only to prevent the enemy from marching on Verdun, which contradicted the offensive mission it had been given by 3rd Army.
It is curious that Engerand argues that Ruffey intended to use reserve divisions in an offensive role, while so many other French apologists claim that the French were beaten solely because they had not planned to use reserve divisions in anything other than secondary roles.
In any case, the leading elements of the of 54 DR were 15km behind VI CA; it was unlikely that 67 DR, which was even further to the rear, could have reached the battlefield on 22 August. Even if the two reserve divisions could have attacked, they would have run into the German 33 ID, supported by a cavalry division, which was waiting for just such a French operation, and the outcome for the French reserve divisions would not have been good.
23 August 1914
At 0730 on 23 August, Joffre sent his evaluation of the situation to the Minister of War, Messimy. He divided the front in half at Virton – Longwy. To the right of that line, the French were advancing slowly, although they enjoyed a significant numerical superiority. The French artillery had silenced the German artillery. On the left of this line the terrain was unfavourable and the French advance had met with ‘great difficulties’ in spite of a considerable numerical superiority. The French were attacking the Germans who were still in march column and the Germans must also be in a difficult situation. The French task was to continue the fight, utilizing their numerical superiority.
The 1800 intelligence summary said that GQG had no information concerning enemy forces on the 4th Army front. 3rd Army had made contact with the German VIII and XVI AK and elements of XIII and XXI AK. This report was mostly wrong: VIII and XXI AK were nowhere near the French 3rd Army sector, but V AK,VI AK and VI RK were.
GQG was completely out of touch with the real situation on the 3rd and 4th Army sectors. Had these armies been dependant on guidance from GQG, they would have been destroyed. The French troops, however, had taken matters in their own hands and as a matter of simple self-preservation had retreated out of the range of the German armies.
French 3rd Army
Ruffy’s orders for the night of 22–23 August instructed the 3rd Army to defend north of the Chiers. ‘The enemy had not followed V CA and had suffered as badly as we had’. At 0030 he ordered 3rd Army to resume the offensive on the morning of 23 August. When GQG expressed the desire at 0245 to have 3rd Army resume the attack, Ruffey responded that he had already given the orders to do so. At 0810 3rd Army submitted a situation report that showed how little the army HQ knew. It said that 40 DI had withdrawn to Nouillonpont; some elements had ‘arrived’ in Etain. IV CA was in good shape. General Trentinian was at the head of 7 DI (not dead), of which two regiments had been reduced to a battalion each (a rosy evaluation). There were no precise reports from VI CA, which was ‘strongly established’ on the Crusnes, but when these reports arrived 3rd Army would renew the attack with IV CA and V CA.
Then the reports from the corps began to arrive.VI CA said that its right flank had been threatened with envelopment and it had withdrawn to the Othain.V CA reported that the troops were incapable of conducting a defence and requested to withdraw over the Chiers. At 0930 Ruffey gave up the idea of attacking. Until 0930 on 23 August the 3rd Army HQ was divorced from reality and exercised no control over the situation.
IV CA reported at 1100 that the situation was ‘very good’. Both divisions were reorganising. DI had already reconstituted five battalions (of twelve!), 7 DI had six battalions. It is hard to see how divisions that had lost half their infantry could be in good condition.
At 1300, 3rd Army sent a report to GQG: the army’s situation was good. The reconstitution of IV CA and V CA was proceeding well. The enemy had been ‘severely tried’. IV CA was behind the Crusnes to the east of Montmédy and not under pressure. 8 DI had reconstituted itself reasonably well, 7 DI had been ‘severely tested by the preceding day’s combat’. V CA was on the Othain to the west and south of Longuyon; most of the corps had been ‘severely tested’. VI CA was just north of the Othain, to the south of Longuyon. It too had been ‘severely tested’ but was ‘preparing to resume the offensive’. ‘The debris of the 40 DI were in reserve at Pillon (8km due south of Longuyon) ready to counterattack.’ 54 DR was supporting the VI CA right flank and ‘the counterattack by 67 DR, 73 DR and 75 DR was being prepared under favourable conditions’
But then harsh reality set in. In a telephonic report that the army gave GQG at 1330, 3rd Army said that ‘the debris of V CA was conducting a major withdrawal to the Loison’. The artillery was in good condition, but the infantry was no longer able to hold its ground due to casualties and fatigue. The enemy was not pursuing vigorously.
The 3rd Army intelligence summary sent in at 1700 shows that the army literally did not know what had hit it. It said that it had identified in its sector the German XVI AK, XVI RK, VI RK, XXI AK and elements of XIII and XII AK. This was mostly wrong. XVI RK did not exist; XII AK and XXI AK were not in the 3rd Army sector. The army had failed to identify V AK,VI AK and V RK.
French 4th Army
At 0500 on 23 August Langle ordered the 4th Army to withdraw to the Chiers, covered by 2 DIC. He reported to GQG that the morale in XVII CA was not good. Its retreat had forced XI CA to withdraw to the Semois, which was not an obstacle to a German attack. Both corps needed to be pulled out of the woods and into open terrain. XII CA had both flanks in the air and had to fall back to Florenville, but it and the Colonial Corps would also have to move even further back, and the whole army might well have to retreat to the Chiers or the Meuse.
Joffre replied at 0830 that according to the reports he had received, there were only three German corps on the 4th Army front. He told Langle to resume the offensive as soon as possible. Joffre was absolutely correct; on 22 August the French 4th Army had faced only three corps, and one of those was a reserve corps. Langle issued a pro-forma order for IX, XI and XII CA to resume the advance. Nevertheless, the 4th Army had been so badly beaten up that it was incapable of doing anything other than retreat.
At 1900 on 23 August Joffre reported to the War Minister that ‘the offensive between Longwy and the Meuse had stopped momentarily.’ This was due to the failure of several individuals who would be dealt with. Three divisions had taken particularly heavy losses. Joffre said he would attempt to resume the offensive.
Four French armies had taken the offensive and been badly beaten. Nevertheless, Joffre’s official explanation was that there were no systematic problems; all that was required was to eliminate some incompetent officers. Joffre had not mentioned that on 24 August the 3rd and 4th Armies would begin to withdraw to the Meuse.
German 5th Army
During the night of 22–23 August AOK 5 received reports from the corps which gave a picture of complete French defeat: the French units had been broken up, the French troops were retreating in great disorder, and French command and control appeared to have collapsed. AOK 4 reported that its fight at Rossignol and Tintigny was ‘not unfavourable’. A LNO sent to XVI AK reported that the French had been completely routed and that there were no signs of French troops west of Metz. Wilhelm wrote that these were ‘unforgettable hours’.
At 0600 5th Army received an operations order from OHL granting the army full freedom of movement, that is to say, the army no longer was required to maintain contact with Metz-Diedenhofen. OHL wanted the left wing of 5th Army – XVI AK – to push the French right wing to the north, away from Verdun. The AOK 5 operations order at 0625 23 August said that the French were in ‘panic flight to the Meuse’ and directed the corps to ‘energetically pursue the enemy, push him away from Verdun and transform yesterday’s victory into a catastrophe’. At 0730 AOK 5 reported to OHL a ‘Complete victory yesterday to the south of Longwy. I am pursuing…’
However, Wilhelm admitted that there was no effective pursuit by 5th Army on 23 August. The troops had been further exhausted constructing defensive positions during the night. The loss of so many leaders necessitated extensive reorganisation and this consumed time. The tactical leadership waited for reconnaissance reports to clarify the situation. Once that was accomplished and the advance began, it was stalled by French rear guards. The terrain gained was insignificant.
In the V AK sector, 9 ID defended in place and 10 ID gained little ground, but XIII AK attacked early in the morning and made the most progress, gaining 15km, an indicator of what it might have accomplished on the afternoon of 22 August. During the day it appeared that the French were going to make a stand on the Othain. The spook of a French counterattack from Verdun materialised again, as French columns were seen advancing towards Spincourt and Etain. The XVI AK commander stopped the corps at 1330, before it entered Spincourt, and began orienting it, as well as the 6 KD and two newly arrived Landwehr brigades, to the south, ending the possibility that the 5th Army would turn the French right flank.
German 4th Army
At 2040 on 22 August, 4th Army issued an operations order renewing the attack at 0500 the next morning with all five corps. At 0630 a similar order arrived from OHL. Air reconnaissance revealed that the French were withdrawing in good order to the Semois. On the army right flank VIII AK, which had hardly been engaged on 22 August, gained only 6km against French forces that had recently arrived. VIII RK attacked, supported by 21 ID, but 25 ID did not move at all. In the XVIII RK sector, 21 RD moved out only at 1200, 25 RD at 1400.VI AK began to advance at 0900, made contact with the 2 DIC, which was still relatively undamaged, and gained only 2km. Due to the heat and the exhaustion of the troops caused by combat and forced marches on 22 August, the 4th Army pursuit produced no results other than occupying terrain. At 1425 AOK 4 reported to OHL:
‘We were completely victorious! We have taken thousands of prisoners, including general officers and many guns. Army is in pursuit of the beaten enemy …The troops fought wonderfully. Many units had very heavy casualties.’
French 4th Army
4th Army reported at 0200 on 24 August that it had another hard day on 23 August. The sole remaining units of the Colonial Corps, 2 DIC and the corps artillery, had been attacked and driven out of their first position. In the afternoon, XII CA on the army left flank had been attacked and forced to retreat, which caused the Colonial Corps to abandon its second position. XII CA reported at 1800 that it needed to reorganise on 24 August and was unable to take the offensive. The corps also required several days rest and replacement officers and NCOs. IX CA on the army left, which had not fought on 22 August, reported at 2000 that it had been pushed back late that afternoon. The 33rd Brigade had ‘preserved its cohesion’ and the artillery had suffered few casualties but the 36th Brigade was no longer combat-effective. Both 136 RI and 77 RI had lost a thousand men and their morale was poor. 4th Army said that ‘under these conditions, the 4th Army offensive found itself temporarily suspended’. Langle reiterated that it would be necessary to fall back to the Chiers and Meuse.
The French View
Engerand’s book Bataille de la Frontières, published in 1920 on the basis of French primary sources, is broadly representative of the French view of the battle. He was critical of Joffre’s offensive strategy and therefore looked kindly on many of the generals Joffre relieved, including Ruffey, the 3rd Army commander and Trentinian, the 7 DI commander. According to Engerand, the 4th Army pushed into the dense forest of the Ardennes without conducting proper reconnaissance and ran into German forces ‘completely dug-in’. The French persisted in senseless attacks until the Germans counterattacked and drove them back. Engerand was, however, principally concerned with the 3rd Army sector. Engerand said that V CA in the 3rd Army centre launched three successive attacks, without artillery preparation, against German entrenchments, decimating three regiments. The corps commander, demoralised, ordered a retreat and was relieved (by Ruffy, presumably) in favour of his chief of staff, who stopped the retreat ‘and led the troops back to their positions’. The V CA commander also failed to cover the right flank of 7 DI on his left. Fortunately, Trentinian sent out a flank-guard detachment (II/101) and also personally ensured that his advanced-guard held the vital strongpoint at Éthe until 7 DI withdrew from the town on orders. The advance guard of 8 DI, 130 RI, ran into German troops which had been able to construct a defensive position just to the north of Virton without being noticed. 130 RI advanced ‘boldly’ and the division deployed to prevent being enveloped. When the fog lifted, the German heavy artillery forced 8 DI to abandon Virton. The IV CA commander massed the corps artillery, and the French 75s stopped the German advance. With their support, the French infantry retook Virton and an attack by 117 RI threw the Germans back into the woods. Engerand, like Trentinian, maintained that the withdrawal of V CA prevented IV CA from exploiting its victories at Éthe and Virton.
Langle, the commander of the 4th Army, writing in the mid-1930s, demonstrated a similar capacity for self-deception. He said that the Germans had been fortifying their positions since 20 August and were fully prepared to receive the French attack. II CA had its march delayed and exposed the flank of the Colonial Corps. The Colonial Corps had been engaged imprudently, with insufficient reconnaissance to the right and had suffered a ‘very serious reverse’. According to Langle, XII CA reached its objectives. XVII CA had encountered an entrenched and invisible enemy and suffered the same fate as the Colonial Corps, for the same reasons, although taken somewhat fewer casualties. The XI CA had pushed back the enemy but had sustained ‘appreciable losses’. The Germans had taken as many casualties as the French 4th Army and had not pursued. Langle said that the 4th Army had not been seriously hurt and remained capable of continuing its mission. XII CA made a successful counterattack on the afternoon of 23 August. 4th Army withdrew on 24 August only because 5th Army on the left and 3rd Army on the right were withdrawing. Langle said that the real problem was Joffre’s insistence that the enemy be attacked wherever he was found. This led to precipitate action. Langle maintained that he was an advocate of methodical, secure operations.
Langle was unable to distinguish between an attack on a prepared position and a meeting engagement. His repeated assertions that the Germans had been digging in for two days betray his complete ignorance of the German situation – 15 years after the battle. At the operational level, the French 3rd and 4th Armies did not have the luxury of a Langle’s slow, methodical advance, which would have negated Joffre’s strategy, indeed negated the entire Russo-French offensive strategy, as well as giving the German right wing the time to appear in the two armies’ rear. Langle’s complaint that the corps columns did not coordinate with each other is misplaced: ensuring such coordination was the job of the 4th Army HQ. On a tactical level, the 3rd Colonial Division and 33 DI were not destroyed because they were advancing rashly, but because the Germans counter-reconnaissance had blinded the French patrols, and the Germans manoeuvred at a rate of speed that befuddled the French division commanders.
Bastin says that at Bellefontaine 120 RI took heavy casualties because the German IR 38 was ‘well-entrenched’. In reality, it was no such thing. He also mentions French bayonet charges three times. In all three cases it is highly unlikely that any such thing occurred; French bayonet charges are never mentioned in the German sources.
The fascination, common to almost all French soldiers and historians, with German trenches and French bayonet charges has nothing to do with actual combat. It was a means of explaining French defeat that emphasised French heroism and avoided confronting German tactical superiority. For modern historians, German trenches and French bayonet charges provide exactly the correct explanation for French defeat, one that corresponds with the popular ‘heroes led by donkeys’ thesis, as well as the experience of the next four years of trench warfare.